AI Ethics Education: A Systematic Literature Review

Abstract

The potential of AI technology to transform human life, wellbeing, and daily work, is faced with risks and challenges yet to be fully accounted for. Scholars and professionals minimize the negative impacts of AI by finding effective practices that implement ethical principles, and in turn, promote the responsible use and design of AI. Education, here, is the practice in focus. However, the complexity of AI ethics makes it hard to pin down what to teach, how to teach it, and how to assess its effectiveness. This paper presents a systematic literature review and in-depth qualitative analysis on the early years of AI ethics education to analyze whether its future trajectory is supported by educational best practices. Our review highlights core challenges in AI ethics education and the content, assessment, and pedagogy used in real interventions. We found this field conceives of AI ethics by teaching from a holistic view (as opposed to a narrow view) through case studies and group projects that challenge students’ ethical reasoning skills in applied practices. However, many papers did not use justified assessment techniques that support student learning, rather, assessment was conducted for research evaluative purposes. This gap in assessment raises implications for researchers and practitioners, as defining success metrics for responsible AI and translating it educational assessment is an ongoing challenge. All in all, our educationally-minded review of AI ethics education details the trajectory of the field to further translate AI ethics principles into distributed ethical practices via formal and informal educational efforts.